Heart failure (HF) is designated as an emerging epidemic. Yet, it is not fully characterized. Most data, derived from hospital discharges, cannot measure incidence, have uncertain validity and cannot capture the full burden of HF because of the shift towards outpatient care. Regarding its etiology, the respective role of hypertension and coronary heart disease (CHD) is controversial. Moreover, the prevalence of obesity and diabetes mellitus is increasing, both conditions linked to HF via several mechanisms such that their contribution to HF could conceivably be increasing but remains to be examined. Finally, while the existence of diastolic HF is recognized, its diagnosis is exclusionary based on symptoms of HF in the absence of LV systolic dysfunction. This approach is unsatisfactory, thus the contribution of DHF to HF remains contentious. These striking gaps in knowledge underscore the necessity of a rigorous investigation of the HF epidemic. Through surveillance of the Olmsted County community, we demonstrated the postponement of CHD towards older ages and the decline over time in the severity of hospitalized MI and the incidence of HF after MI. This implies that, if CHD is the main cause of HF, HF should be postponed towards older ages and its incidence rate relatively stable. During the same period, preliminary findings on HF surveillance suggest that the incidence of first clinical diagnosis of HF may not be increasing as much as implied by hospital discharges and that adverse trends may be occurring preferentially among younger ages. These data from the same community are challenging to reconcile with the concept of an ongoing major contribution of CHD to an epidemic of HF, thereby underscoring the need to rigorously study the epidemiology of HF, which is the focus of this application. We propose 3 specific aims and a community surveillance approach, integrated with our ongoing work on CHD surveillance to investigate the HF epidemic in Olmsted County by characterizing its magnitude and determinants and studying prospectively the contribution of DHF. Aim 1 will estimate the secular trends in the incidence and in the outcome of validated HF to test the hypotheses that there has been an increase in the incidence of HF, which differs by age and sex and that the survival of HF improved while hospitalization for HF has increased. Aim 2 will use a case-control approach to characterize the etiology of HF and its changes over time to test the hypotheses that CHD and hypertension confer an excess risk of HF, the magnitude of which is declining over time, that obesity and diabetes mellitus confer an excess risk of HF the magnitude of which is increasing and that the population attributable risk of CHD and hypertension for HF is declining, while that of obesity and diabetes mellitus is increasing over time. Aim 3 will prospectively characterize the contribution of DHF to HF using brain natriuretic peptide (BNP) among persons with HF and define the prognostic value of BNP in all cases of HF. Thus, the completion of these aims will provide important insights into the epidemiology of HF.